Robertuito Sentiment Analysis

🚀 Sentiment Analysis in Spanish

This project provides a sentiment analysis model for Spanish text, trained on the TASS 2020 corpus using the RoBERTuito base model.

🚀 Quick Start

This repository offers a sentiment analysis model for Spanish. The model is trained with the TASS 2020 corpus, which contains around ~5k tweets in various Spanish dialects. The base model is RoBERTuito, a RoBERTa model pre - trained on Spanish tweets.

Repository: https://github.com/pysentimiento/pysentimiento/

✨ Features

• Trained on Diverse Spanish Dialects: Utilizes the TASS 2020 corpus, covering multiple Spanish dialects.
• Standard Labels: Uses POS, NEG, NEU labels for sentiment classification.

📦 Installation

To use this model, you need to install the pysentimiento library. You can install it via pip:

pip install pysentimiento

💻 Usage Examples

Basic Usage

You can use the model directly with pysentimiento:

from pysentimiento import create_analyzer
analyzer = create_analyzer(task="sentiment", lang="es")

analyzer.predict("Qué gran jugador es Messi")
# returns AnalyzerOutput(output=POS, probas={POS: 0.998, NEG: 0.002, NEU: 0.000})

📚 Documentation

Results

The following table shows the Macro F1 scores for the four tasks evaluated in pysentimiento:

Note that for Hate Speech, these are the results for Semeval 2019, Task 5 Subtask B.

📄 License

If you use this model in your research, please cite pysentimiento, RoBERTuito and TASS papers:

@article{perez2021pysentimiento,
  title={pysentimiento: a python toolkit for opinion mining and social NLP tasks},
  author={P{\'e}rez, Juan Manuel and Rajngewerc, Mariela and Giudici, Juan Carlos and Furman, Dami{\'a}n A and Luque, Franco and Alemany, Laura Alonso and Mart{\'\i}nez, Mar{\'\i}a Vanina},
  journal={arXiv preprint arXiv:2106.09462},
  year={2021}
}

@inproceedings{perez-etal-2022-robertuito,
    title = "{R}o{BERT}uito: a pre-trained language model for social media text in {S}panish",
    author = "P{\'e}rez, Juan Manuel  and
      Furman, Dami{\'a}n Ariel  and
      Alonso Alemany, Laura  and
      Luque, Franco M.",
    booktitle = "Proceedings of the Thirteenth Language Resources and Evaluation Conference",
    month = jun,
    year = "2022",
    address = "Marseille, France",
    publisher = "European Language Resources Association",
    url = "https://aclanthology.org/2022.lrec-1.785",
    pages = "7235--7243",
    abstract = "Since BERT appeared, Transformer language models and transfer learning have become state-of-the-art for natural language processing tasks. Recently, some works geared towards pre-training specially-crafted models for particular domains, such as scientific papers, medical documents, user-generated texts, among others. These domain-specific models have been shown to improve performance significantly in most tasks; however, for languages other than English, such models are not widely available. In this work, we present RoBERTuito, a pre-trained language model for user-generated text in Spanish, trained on over 500 million tweets. Experiments on a benchmark of tasks involving user-generated text showed that RoBERTuito outperformed other pre-trained language models in Spanish. In addition to this, our model has some cross-lingual abilities, achieving top results for English-Spanish tasks of the Linguistic Code-Switching Evaluation benchmark (LinCE) and also competitive performance against monolingual models in English Twitter tasks. To facilitate further research, we make RoBERTuito publicly available at the HuggingFace model hub together with the dataset used to pre-train it.",
}

@inproceedings{garcia2020overview,
  title={Overview of TASS 2020: Introducing emotion detection},
  author={Garc{\'\i}a-Vega, Manuel and D{\'\i}az-Galiano, MC and Garc{\'\i}a-Cumbreras, MA and Del Arco, FMP and Montejo-R{\'a}ez, A and Jim{\'e}nez-Zafra, SM and Mart{\'\i}nez C{\'a}mara, E and Aguilar, CA and Cabezudo, MAS and Chiruzzo, L and others},
  booktitle={Proceedings of the Iberian Languages Evaluation Forum (IberLEF 2020) Co-Located with 36th Conference of the Spanish Society for Natural Language Processing (SEPLN 2020), M{\'a}laga, Spain},
  pages={163--170},
  year={2020}
}